Or use an RM core with a hole through the middle. Thick wire wound conventionally, sense wire uses turns through the hole. Should reduce coupling but still allow core saturation sensing.
Where is the current sensed?
In this case the reflected impedance is sensed, not current. The effect remains the same: an apparent inductance loss and hence the reactance loss. If you replace the switch sensing winding with a current carrying conductor, you'll have your current indicator. Except for the L and R6, everything else can be pushed into an MCU if you happen to have one around.
Best regards, Piotr
1- 1/2" wide .005 steel strip can be retrofitted (bend clip over cable). 2- Silicon steel strip is free from garbage, sensor from any old PC fan. 3 - it has minimal bulk or weight. 4 - adjustable, on the spot (though permanent fixation is required for repeatable output) 5 - low power, 3-wire interface. 6 - no research required, but you may spend a lot of time pfaphing about, getting it to go. (well, the sensor has to be working . . .)
RL
That circuit senses impedance, not current. It could have zero current and any amount of apparent coupling.
A real flux-gate type sensor can't be a simple transformer, namely two windings on one magnetic core, because that becomes an impedance sensor. Even a second harmonic is affected by loop impedance.
It take two cores in a tricky configuration (I've done that) or a crossed-field thing.
The basic shunt circuit is simple. One could add adjustable threshold, hysteresis, and load dump protection as needed.
It's almost embarassing to post something this simple.
Pathetic evasion.
If it looks like a duck, swims like a duck, and quacks like a duck, then it probably is a duck. The same applies to stupid morons.
It's often asserted that "Boris is a very clever man", usually by those who wouldn't recognise a clever man if it bit them in the arse. A classical Eton and Oxford education is no guarantee of intelligence, it just opens the right doors.
This current Johnson government, made up as it is from one-trick-pony brexiters, has plumbed the depths of ineptitude. It wouldn't be so bad if it hadn't resulted in the death of almost 128,000 people here in the UK.
John has missed the point. The device is sensing core saturation, and the reactive impedance is the proxy for core saturation
It isn't any kind of transformer. There's no magnetic coupling between the sense loop and the coil that carries the current being sensed. The loop running through the centre hole in the RM core is - in theory - coupled to the coil on the RM former but the flux generated by that coil is almost completely confined within the RM core, so the net flux through the search coil doesn't change.
It is a crossed filed thing. <snip>
Isn't it just.
I'm not arguing that Boris is competent or realistic. The problem with people like Donald Trump and Boris Johnson is that they devote their considerable intelligence to telling people convincing stuff that makes the people think well of them, even though it is wrong and misleading and like to have bad long term results.
Intelligent bull-shit artists do more damage than dumb ones because they can fool more people for longer.
False. I'm discussing the simplicity issue, not just the circuit design. I hear the dismissive "it sounds complicated" as an excuse to snub a perfectly sound engineering solution.
A long discussion of resistivity can make one certain that 'it's complicated' applies to every resistor. And, in some circumstances, it IS complicated. You can design around that, though. Don't snub 'complicated' resistors. Or inductors. Nonlinear magnetics are useful. Heck, carbon resistors' conductance drop is the basis of a lot of liquid nitrogen level sensors. I built a capacitive dipstick type, but no one cared because the other solution is easier.
Pathetic snipping.
The whold thing could be done on a scrap of copperclad FR4. The shunt resistor could be dremeled into the top copper in 30 seconds, by dead reckoning or fine trimmed if one cares to do that.
I posted the basic circuit.
Post the circuit of your magnetic concept.
Crudley, but simply enough for John Larkin to understand it.
Don't bother. The various ways of sensing the current by getting it to set up a magnetic field have been spelled out here already, and John Larkin doesn't like any of them, and doesn't seem to even understand what would be going on in the variations where the current would be used to saturate a lump of ferrite.
EIGHT magnetic components!
My shunt circuit has four parts, three if you dremel the shunt on the pcb.
Yeah, but that bunch of magnetic parts gave safe isolation of a high-voltage AC power system, with high-bandwidth current sense output. An automobile battery, at 12VDC, with 'did it happen yet' sensing doesn't get the same treatment. And, you are ignoring the how-to-attach issue entirely. Pruning the real problem down to a toy one, does achieve simplicity.
And every single one of them very small, yet super robust. All the cores are 6mm OD F938. The number of turns is 3x20 and 1x33, respectively. The magnetizing windings of the current sensors are the source pins of the TO247 power MOSFETs. This circuit is unbelievably forgiving.
Planet Earth to John: you are comparing a complete current-driven synchronous rectifier to a simplistic current threshold detector. Add some FPGAs to the BOM, because why not?
Best regards, Piotr
Your magnetophobia is getting clinical.
Piotr
I'm old fashioned. 5 cheap standard parts seems better to me than a couple dozen parts, many custom inductors.
Sloman loves magnetics so never finishes anything.
I've done active CTs, DCCTs, kilowatt power transformers, transmission-line transformers, kilovolt air-core inductors, planar transformers, custom toroids, fast gate drive transformers, blocking oscillators, RF magnetics, custom pot cores, rod-core power transfer, pot-core power transfer, all sorts of weird stuff.
But I only use custom magnetics when there's no simpler way. A resistor seems to me to be a pretty good way to measure current.
That's not the issue at hand. Optos are usually better than magnetics for isolation.
Simplicity has its virtues. As in having something done quickly and working first try, exactly as designed.
Are those custom? What did they cost each?
There is much to forgive.
The stated problem was to sense a current on a 12 volt line, and beep if it went below 200 mA.
Nobody seems to want to do it easily.
A flux gate?! Nobody is suggesting measuring the slight difference in magnetizing fields with bipolar excitation. This is not a microtesla-capable magnetometer intended to blow up a ship. Just measure the R_L(I) with a known waveform.
Of course it can. If you want to decouple the coils, just use two toroidal cores with the windings connected in series or build a composite structure on an E core. Put the sense coils on the outer legs and the magnetizing winding on the center one. This is Magamp design 101.
What is tricky in putting two doughnuts on a rope instead of just one? The crossed-field thing is a core with an additional hole; an RM or P for instance. Perhaps several of them can still be excavated from the cornfield in Roswell.
Same here. In both cases you end up with a current to voltage converter of some shape or form. Both circuits have unique and important properties.
Best regards, Piotr
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.